AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Self-Supervised Aggregation of Diverse Experts for Test-Agnostic Long-Tailed Recognition

Yifan Zhang, Bryan Hooi, Lanqing HONG, Jiashi Feng

NeurIPS 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Decoupled and calibrated long-tail training · 55 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use skill-diverse experts trained with cross-entropy, balanced softmax, and inverse softmaxSADE · ImageNet-LT, uniform testtop-1 accuracy (All): 56.3 -> 58.8Improved
Add test-time self-supervised expert aggregation (ablation)SADE · ImageNet-LT, Forward-LT-50top-1 accuracy (All): 65.5 -> 69.4 (+3.9)Improved
Add test-time self-supervised expert aggregation (ablation)SADE · ImageNet-LT, Uniformtop-1 accuracy (All): 58.8 -> 58.8 (+0.0)No clear effect
Add test-time self-supervised expert aggregation (ablation)SADE · ImageNet-LT, Backward-LT-50top-1 accuracy (All): 49.8 -> 53.1 (+3.3)Improved
Add test-time self-supervised expert aggregation (ablation)SADE · ImageNet-LT, only few-shot classestop-1 accuracy: 42.5 -> 53.6; +11.1Improved
Replace learned expert-specific objectives with independently trained cross-entropy experts as in RIDE (ablation)SADE expert ensemble · ImageNet-LT, uniform testtop-1 accuracy (All): 58.8 vs 56.3Hurt
Use prediction-stability aggregation rather than test-time pseudo-labelingSADE · ImageNet-LT, Backward-LT-50top-1 accuracy: 42.4 -> 53.1Improved
Use prediction-stability aggregation rather than entropy minimization with TentSADE · ImageNet-LT, Backward-LT-50top-1 accuracy: 45.7 -> 53.1Improved
Train with stronger RandAugment for 400 epochs (ablation)SADE · ImageNet-LT, uniform testaccuracy: 58.8 -> 61.2Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy